World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018
Steady State Rolling and Dynamic Response of a
Tire at Low Frequency
Md Monir Hossain, Anne Staples, Kuya Takami, Tomonari Furukawa
Abstract—Tire noise has a significant impact on ride quality
and vehicle interior comfort, even at low frequency. Reduction of
tire noise is especially important due to strict state and federal
environmental regulations. The primary sources of tire noise are the
low frequency structure-borne noise and the noise that originates from
the release of trapped air between the tire tread and road surface
during each revolution of the tire. The frequency response of the tire
Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852
changes at low and high frequency. At low frequency, the tension
and bending moment become dominant, while the internal structure
and local deformation become dominant at higher frequencies. Here,
we analyze tire response in terms of deformation and rolling velocity
at low revolution frequency. An Abaqus FEA finite element model
is used to calculate the static and dynamic response of a rolling tire
under different rolling conditions. The natural frequencies and mode
shapes of a deformed tire are calculated with the FEA package where
the subspace-based steady state dynamic analysis calculates dynamic
response of tire subjected to harmonic excitation. The analysis was
conducted on the dynamic response at the road (contact point of tire
and road surface) and side nodes of a static and rolling tire when
the tire was excited with 200 N vertical load for a frequency ranging
from 20 to 200 Hz. The results show that frequency has little effect on
tire deformation up to 80 Hz. But between 80 and 200 Hz, the radial
and lateral components of displacement of the road and side nodes
exhibited significant oscillation. For the static analysis, the fluctuation
was sharp and frequent and decreased with frequency. In contrast, the
fluctuation was periodic in nature for the dynamic response of the
rolling tire. In addition to the dynamic analysis, a steady state rolling
analysis was also performed on the tire traveling at ground velocity
with a constant angular motion. The purpose of the computation
was to demonstrate the effect of rotating motion on deformation and
rolling velocity with respect to a fixed Newtonian reference point.
The analysis showed a significant variation in deformation and rolling
velocity due to centrifugal and Coriolis acceleration with respect to
a fixed Newtonian point on ground.
Keywords—Natural frequency, rotational motion, steady state
rolling, subspace-based steady state dynamic analysis.
I. I NTRODUCTION
N
OISE has an adverse environmental effect on urban
life. Though there has been tremendous development
in automotive technology in recent years, noise generated
from vehicles is still a big concern for the auto industry for
environmental reasons and because it impacts ride quality.
Many moving parts of vehicle generate noise. A significant
portion of the noise is generated by the tire/road interaction.
The two main sources of tire noise are 1) structure-borne noise
and 2) the noise generated by the horning effect in the tire
grooves during tire /road interaction [1].
In the horning effect, air is sucked in and pushed out of the
contact area at the tire tread during each tire revolution, which
Md Monir Hossain is with the Virginia Tech, United States (e-mail:
mdmoh81@vt.edu).
International Scholarly and Scientific Research & Innovation 12(4) 2018
creates an air pumping effect, which generates high frequency
noise [2]. In the case of tires without the block pattern found
in current tire models, the main source of noise is restricted
to structural-borne noise.
Previous experiments and simulations have mostly
concentrated on eliminating high frequency noise (above 200
Hz) [3]. But the behavior of tire noise varies substantially
with frequency due to the complex structure of tires. For
structural noise, tension and bending becomes predominant
at low frequency and torsion becomes important at higher
frequency [4]. So the radial and lateral deformation of any
point on tire is not similar at low and higher frequency range.
It is also convenient to follow the noise at higher frequency.
But at lower frequency, the dynamic response of tire is
not easily tractable. For this reason, the current paper only
investigate the effects of low frequency noise (below 200Hz)
on tire deformation and rolling motion.
To understand the effect of low noise, employing correct
prototypes of the tire is very important. To reduce the
complexity of analysis, ring or plate/shell shape analytical
models [5]-[7] were proposed earlier. These simplifications
have reduced the computational effort but limited the
importance of complex structural configuration of modern tire.
Now for simulation, similar simplified models have been used
earlier and only radial part of contact force was considered
by neglecting tangential force of moving tire . To remove the
constrained mentioned above, we need a proper FE model
which can incorporate correct material behavior and forces
developed during tire rotation. This detailed finite element
models is also necessary to study dynamic response of a tire at
different frequency through Eigen value analysis [8]. Now the
dynamic response of tire is not similar for static and rolling tire
[9]. But most of the previous analysis have been conducted on
static tire and neglected centrifugal, Coriolis effects and the
dependency of Eigen frequency on rolling motion. So it is
necessary to study and compare the dynamic response of a
static and rolling tire at low frequency range.
Now for the Eigen value analysis, the cost depends on
matrices provided by the finite element model and large-scale
systems of linear equations. Generally direct solution analysis
is used to calculate linearized response of the dynamic system.
Though the result from the analysis is very accurate, the
method is computationally expensive for complex FE model
of tire [10]. So we need an alternative method which will
solve the problem with almost similar accuracy but at less
computational cost. The subspace-based steady state dynamics
is such an effective solution for the problem mentioned above.
In this method, steady-state dynamic equations are projected
416 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018
on a subspace of selected modes of the undamped system for a
frequency range and the system is excited at Eigen frequency
extracted prior to such analysis. The projection of dynamic
equilibrium equations into a subspace of selected modes
reduce the system of complex equations and computational
cost.
Static and steady state transport analysis have to be
performed prior to dynamic analysis. Modeling rolling motion
of a deformable body, in contact with a surface is difficult
for Lagrangian analysis. Because the reference frame is in
motion and for observer, the problem becomes transient. But
if we are able to attach the reference frame to the axle of
rolling tire, the motion of any particle of tire will be fixed
with respect to the observer on axle but will only move with
deformation of tire. This optimal approach is known as mixed
Eulerian-Lagrangian method in which the rigid body rotation
Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852
is described in Eulerian (spatial) manner and the deformation
is described in Lagrangian manner. A short overview on steady
state rolling tire analysis can be found in literature review
[11], [12]. Now the problem arise when rotational motion is
introduced in analysis. Rotational motion enforces centrifugal
and Coriolis acceleration which were absent in most of the
previous computations. Besides, velocity and acceleration of
any point on tire will not be similar with respect to reference
point on tire and fixed Newtonian reference point on ground.
Here the tire is not only rotating on its own reference axis but
also rotating against a fixed reference frame on ground. So the
calculation of any parameter with respect to fixed reference
point on ground should include change of local parameter on
rotational body and relative motion of local reference point
with respect to fixed point on ground.
To remove the constraints discussed above, first part of
the current paper deals with steady state analysis of a
rolling tire with respect to fixed Newtonian fixed point on
ground.As a consequence,centrifugal and Coriolis force have
been included in the Eulerian-Lagrangian analysis. Before
steady state rolling, static foot print analysis of the tire has
been performed with different loading conditions and the
corresponding deformations have been analyzed for half and
full tire model The second part of the paper deals with dynamic
analysis of tire. At first, dynamic analysis has been conducted
on static tire and the natural frequencies and mode shapes of
a deformed tire have been calculated by using finite element
package (Abaqus). Then the FE model of tire was excited
with 200N harmonic load at low frequency (20-200 Hz) and
the corresponding response of road and side node of tire
have been obtained at low frequency range. For rolling tire,
similar analysis have been performed and the results have been
compared with static tire analysis. The results of current study
will help us to understand steady state and dynamic response
of a static and rolling tire at low frequency range.
II. M ODEL G ENERATION
For correct computational analysis, the elements of FE
model should be selected carefully. The tire model can be
either discretized with equal number of elements or can be
discretized non uniformly with more elements near the region
International Scholarly and Scientific Research & Innovation 12(4) 2018
Fig. 1 FE partial model of a tire
of interest. The non uniform distribution of elements makes
the computation faster and more elements near the region of
interest reduce computational error. Couple of test analysis
have been performed for both uniform and non uniform mesh
models to check the accuracy of the results and the out come
of computations were consistent and similar for both the
cases. The axisymmetric model was discretized with bilinear
elements. A partial three-dimensional model was generated by
revolving the axisymmetric model about the rotational axis.
The generated partial model is shown in figure 1. This partial
three-dimensional model is also composed of bilinear elements
except for footprint region where more linear elements have
been utilized to increase resolution near contact patch. The
full three-dimensional model was generated by reflecting the
partial three-dimensional model. One thing should be noted
that all the elements used for the analysis is hybrid in nature.
The hybrid element is suitable for large deformation and avoid
shear and volume locking. The construction of FE model
was very similar to real tire except for the absence of any
specific tread pattern. The tread pattern was not introduce in
the model to restrict the source of noise only from structure
of tire. The tread and sidewalls are made of rubber, and
the belts and carcass are constructed from fiber-reinforced
rubber composites. The viscoelastic rubber was modeled as an
incompressible hyper-elastic material by 1-term Prony series.
For viscoelastic model, the value of relaxation modulus was
g p = 0.3 and its associated relaxation time was τ p = 0.1.
For accurate representation of FE model, property of material
should be similar to real tire values.
III. S TATIC A NALYSIS OF T IRE
Static analysis of a tire was performed prior to steady state
rolling analysis. Then the final results from static analysis
was transferred as the initial results for steady state analysis.
During static analysis,tire was kept in static condition and
load was applied to the tire over several steps. In the current
simulation, the tire was inflated with a pressure of 200.0 kPa
and a vertical load of 1.7 kN was applied to the rigid body
reference node for partial three-dimensional model to represent
a 3.4 kN load in the full three-dimensional model. The
contact between road and tire was modeled as hard contact.
The penalty method approximates hard pressure-overclosure
behavior. As the tire was static, the friction between road and
tire was considered zero and the contact between two surfaces
were modeled using contact pair algorithm. The deformation
of final step is shown in Fig. 2 and contact area and pressure
is shown in Fig. 3.
417 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018

Fig. 2 Deformation of static tire (a) Deformation of equal grid model

Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852

(b) Deformation of full tire model

(a) Contact area Fig. 4 Deformation of half and full static tire model

(b) Contact pressure

Fig. 3 Contact area and pressure of a static tire at footprint region Fig. 5 Steady state analysis model

In the previous case, the footprint region was discretized
with more elements than the remaining circumferential
element of tire. In the second case, the tire was modeled
with equal number of bilinear elements over the partial tire
model. For the second simulation, similar steps were followed
as discussed in previous case with identical applied loads and
boundary conditions and the interaction region between road
and tire displayed almost similar deformation. Fig. 4 (a) shows
the component wise deformation of partial model tire. Once
the static footprint solution for the partial three-dimensional
model has been established, symmetric results transfer was
used to transfer the solution to the full three-dimensional
model and the footprint solution was brought into equilibrium
in a single static increment. The deformation of full tire is
shown in Fig. 4 (b).
IV. S TEADY S TATE T RANSPORT A NALYSIS
For steady state analysis in current study, a mixed Eulerian/
Lagrangian method has been used where the rigid body
rotation is describer in an Eulerian manner and deformation of
tire as an Lagrangian manner. Thus the kinematics of problem
has been converted to transient to spatial dependent problem.
In addition, the analysis requires less time due to fine meshing
only in the region of interest. In the analysis, frictional
effects, inertia effects, and history-dependent material models
were included during the study. The purpose of the steady
state analysis analysis was to obtain free rolling equilibrium
solutions of a 175 SR14 tire traveling at a ground velocity of
10.0 km/h (2.7778 m/s). The model is shown in Fig. 5 [13]
where the ground velocity of body is described in terms of a
constant cornering motion.
For steady state condition, velocity and
acceleration in the reference frame tied to the
body can be expressed by following equations [13]:
∂λ
Vr = Ωn × (x − Xc ) + ωR (1)
∂S
∂X ∂2λ
ar = Ω2 (nn − I).(x − Xc ) + 2ωΩRn × + ω 2 R2 2 (2)
∂S ∂ S
Here Vr and ar is the velocity and acceleration of reference
point on tire respectively. Ω is the rotational velocity with
respect to fixed point on ground and ω is the local angular
velocity. λ is used for mapping between Eulerian and
Lagrangian frame. s represents distance along circumferential
direction of tire and R is the radius of a point on the
reference body. The velocity expression depends only on
local rotational velocity of tire with respect to own axis and
rotational velocity with respect to Newtonian fixed frame on

International Scholarly and Scientific Research & Innovation 12(4) 2018 418 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018
(a) Straight line deformation
Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852
(b) straight line velocity
Fig. 6 Steady state straight line analysis of tire
ground. For acceleration, the first term gives rise to centrifugal
forces. The second term represents acceleration for Coriolis
forces. The last term combines the acceleration for Coriolis
and centrifugal forces resulting from local rotation about own
axis. If the deformation is uniform along the circumferential
direction, this Coriolis effect vanishes so that the acceleration
gives rise to centrifugal forces only. For straight line
motion, as shown in Fig. 5, the first term of velocity can
be neglected. The centrifugal and Coriolis acceleration due
to rotational velocity with respect to Newtonian fixed point
becomes zero . So the final expression for straight line
rolling velocity and acceleration can be expressed as follows:
∂λ
V = V0 + Vr = V0 + ωR
∂S
∂2λ
a = ar = ω 2 R 2
∂2S
Here V0 is the ground velocity at the axle of tire. For straight
line motion, the magnitude of deformation and straight line
velocity is shown in Fig. 6.
During the simulation, only straight line rolling velocity of
tire has been considered and ignored the rotating motion with
respect to fixed Newtonian reference point on ground. Because
of this assumption, centrifugal and Coriolis acceleration due
to rotation with respect to fixed reference point become zero
and the expression becomes very simple which include only
local centrifugal and Coriolis acceleration. Now in the second
case, we have only considered rotating velocity of tire with
respect to fixed reference point on ground and ignored the
local ground velocity. In this case is tire is not only rotating
around fixed Newtonian reference point but also spinning on
its own axis. The velocity profile for this case is shown in
Fig. 7. As we are considering only rotational velocity of the
tire, the magnitude of rolling velocity is less than straight line
International Scholarly and Scientific Research & Innovation 12(4) 2018
(a) Deformation
(b) Rolling velocity
Fig. 7 Deformation and rolling velocity only for rotating motion
(a) Deformation
(b) Rolling velocity
Fig. 8 Linear and rotating velocity with respect to ground fixed point
(3)
(4) velocity case and also there is difference in deformation in
both the cases.
In the third case, both the ground velocity of tire and
rotational velocity with respect to reference point on ground
have been considered. So all the terms of velocity and
acceleration of expression have to be considered in the
computation. The results of rolling velocity and deformation
is shown in Fig. 8. If we compare the three steady state cases,
we will find significant difference in the steady state rolling
velocity and deformation due to change in running parameter.
In case of only linear ground velocity, maximum straight line
velocity was 5.34 m/s. In only rotating velocity case without
ground velocity, the maximum rolling velocity was 2.6 m/s
and in the final case which include both ground velocity and
rotational motion exhibited highest rolling velocity of 5.94
m/s. So there is significant effect of these parameters on tire
rolling motion.
The effect of material property of tire was also studied
during steady state analysis. A simple 1-term Prony series
419 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018

(a) Visco-elastic transient model

(a) Eigen value
Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852

(b) Long term model

Fig. 9 Effect of Material property on deformation of a tire
(b) Eigen frequency
Fig. 11 Eigen value and frequency of static tire
model was used for analysis. In the current analysis, relaxation
modulus g p = 0.3 and its associated relaxation time, τ p = 0.1
have been used for calculation.Time dependent viscoelastic
material property and also the long-term behavior of the
material have been investigated during steady state analysis.
The results of effect of material property on deformation is
shown in Fig. 9. For long term analysis, the deformation was
predicted higher than viscoelastic transient model.
(a) Eigen frequency 42.77 (b) Eigen frequency 72.09

(c) Eigen frequency 113.26 (d) Eigen frequency 176.51

Fig. 12 Mode shapes of tire at different Eigen frequency

(a) Shear stress along the road
shear stress developed due to continuous interaction of tire
with road surface in the footprint region. Fig. 10 shows shear
stress developed near the footprint region along the road and
transverse direction of travel with friction co-efficient 0.2
during straight line motion study.

V. S UBSPACE -BASE DYNAMIC A NALYSIS OF T IRE AT L OW

F REQUENCY

Subspace-based steady-state dynamic study is an analysis

(b) Shear stress transverse of road
Fig. 10 Shear stress in footprint region during steady state analysis
In the static analysis,the value of friction coefficient was
zero and only normal force was acting on contact point at
road node. For the steady state case, the value of friction
co-efficient was not zero and not only contact pressure but
procedure to determine dynamic response of a system
subjected to harmonic excitation by directly projecting the
solution into a reduced-dimensional subspace of undamped
system. Prior to analysis, Eigen frequency has been extracted
within the frequency domain of interest. Figs. 11 and 12
display Eigen value and frequency of FE tire model at different
modes.

International Scholarly and Scientific Research & Innovation 12(4) 2018 420 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018
(a) Deformation (u3) at road node
Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852
(b) Deformation (u3) at side node
(c) Deformation (u2) at side node
(d) Deformation (u1) at side node
Fig. 13 Deformation of static tire at low frequency
The subspace-based steady-state dynamic is a linear method
but nonlinear response from static and steady state rolling
analysis can be included in the study. The static tire was
International Scholarly and Scientific Research & Innovation 12(4) 2018
Fig. 14 Eigen frequency at rolling velocity
excited with a vertical load of 200 N and the frequency was
swept from 20 Hz to 200 Hz. The boundary condition was kept
similar to the static footprint analysis of tire. The objective of
the analysis was to observe frequency response of static tire in
low frequency range at the road node (contact point between
road and tire)and side node (point located on lateral side of
tire).
The response at road node for vertical displacement (U3 )
is shown in Fig. 13 (a). From the plot, it can be observed
that almost nothing happens until 80 Hz. But between 80 Hz
to 200 Hz, there was periodic fluctuation which decreased
with the increasing frequency. The frequency until 80 Hz
was too small to effect the displacement of tire. For side
node, similar frequency response was observed in the vertical
direction but smaller in magnitude. Now for road direction, one
single fluctuation was observed at 60 Hz and the magnitude of
oscillation increased between 80 to 140 Hz. The displacement
magnitude was smaller for road direction with respect to
displacement in other two directions for side node. From the
dynamic response it can be concluded that the displacement
of tire dies with frequency for static tire at lower frequency
range.
The dynamic response of a rolling tire was also studied for
rolling velocity of 60 km/hr. Direct solver was used to solve
the systems of equations. Before the dynamic analysis, real and
complex Eigen values of tire were extracted for rolling motion
to understand the effect of centrifugal and Coriolis forces. Fig.
14 shows the comparison of first 15 pairs of Eigen frequency
between 0 to 200 Hz at two different rolling velocity of 10
and 60 km/hr.
At low rolling velocity, the value of real and complex
Eigen frequency was close in magnitude. But as the velocity
increased, the real and complex Eigen frequency changed and
displayed dependency on rolling motion. The Eigen frequency
was for rolling tire was also different from static analysis.
After calculating the Eigen frequency, the rolling tire was
excited with 200 N vertical load and the response was obtained
at Eigen frequency between 80 to 200 Hz. Displacement for
side node (point located at lateral side of tire)is shown in Fig.
15. In contrast to static analysis the displacement was small
in magnitude and periodic in nature for vertical and lateral
direction and continuously increased between 150 to 200 Hz.
For road direction, the displacement was almost zero until 180
421 ISNI:0000000091950263
 World Academy of Science, Engineering and Technology
International Journal of Aerospace and Mechanical Engineering
Vol:12, No:4, 2018
(a) Deformation (u3) at side node
Open Science Index, Aerospace and Mechanical Engineering Vol:12, No:4, 2018 waset.org/Publication/10008852
(b) Deformation (u2) at side node
(c) Deformation (u1) at side node
Fig. 15 Deformation at side node for rolling tire at low frequency
Hz and then increased very sharply between 190-200 Hz. From
the dynamic response it can be concluded that displacement
becomes small for rolling tire but does not die out at lower
frequency range in contrast to dynamic analysis of static
tire.In contrast to static analysis the displacement was small
in magnitude and periodic in nature for vertical and lateral
direction and continuously increased between 150 to 200 Hz.
For road direction, the displacement was almost zero until 180
Hz and then increased very sharply between 190-200 Hz. From
the dynamic response it can be concluded that displacement
becomes small for rolling tire but does not die out at lower
frequency range in contrast to dynamic analysis of static tire.
VI. C ONCLUSION
Static footprint and steady state transport analysis have
been performed prior to dynamic analysis of tire at different
International Scholarly and Scientific Research & Innovation 12(4) 2018
rolling conditions. For steady state analysis, three cases have
been studied to understand the effect of rotating motion with
respect to fixed Newtonian point on ground. The velocity and
acceleration of reference point on the body were also different
for all the three cases. Then the dynamic response of the
tire was recored for static and rolling tire between 20 Hz
to 200 Hz. The results show difference in Eigen frequency
for static and rolling tire at different rolling conditions. The
results also display that the deformation was negligible until
80 Hz and decreased with increasing frequency between 80 to
200 Hz for static tire. In contrast, the displacement of rolling
tire was small and periodic in nature but did not die out at
lower frequency range. Results obtained in the study will be
helpful to understand steady state rolling motion and dynamic
response of a tire at low frequency range.
R EFERENCES
[1] K. Iwao and I. Yamazaki, “A study on the mechanism of tire/road noise,”
J. SAE Review 17 , vol. 2, 1996, pp.139–144.
[2] S. Kim, W. Jeong, Y. Park and S. Lee, “Prediction method for tire
air-pumping noise using a hybrid technique,” J. Acoustical Society of
America 119, vol. 6, 2006, pp. 3799–3812.
[3] U. Sandberg and J. Ejsmont, Tyre/Road Noise Reference Book. Kisa,
Sweden, 2002.
[4] K. Larsson and W. Kropp, “A high-frequency three-dimensional tyre
model based on two coupled elastic layers,” J. Sound and vibration, 2002,
pp. 889–908.
[5] W. Kropp, K. Larsson, F. Wullens, P. Andersson, F. Becot and T.
Beckenbauer, “The modeling of tire/road noise–a quasi three dimensional
model,” in Proc. Inter–noise, Hague, Netherlands, 2001.
[6] M. Matsubara, D. Tajiri, T. Ise and S. Kawamura, “Vibrational response
analysis of tires using a three-dimensional flexible ring-based model,” J.
of Sound and Vibration 408, 2017, pp.368-382.
[7] W. Kropp, “Structure-borne sound on a smooth tyre,” J. Applied Acoustics
26, vol. 3, 1989, pp. 181-192.
[8] R. Lehoucq, D. Sorensen and C. Yang, Arpack users guide: solution of
large scale eigenvalue problems with implicitly restarted Arnoldi methods,
SIAM, 1998.
[9] C. Diaz, P. Kindt, J. Middelberg, S. Vercammen, C. Thiry, R. Close and
J. Leyssens, “Dynamic behaviour of a rolling tyre: Experimental and
numerical analyses,” J. Sound and Vibration 364, 2016, pp.147-164.
[10] Z. Bai, J. Demmel, J. Dongarra, A. Ruhe and H. Vorst, Templates
for the solution of algebraic eigenvalue problems: a practical guide.
Philadelphia:SIAM, 2000.
[11] U. Nackenhorst, “The ALE-formulation of bodies in rolling
contacttheoretical foundations and finite element approach, Computer
Methods,”J. Applied Mechanics and Engineering, 2004, pp. 4299-4322.
[12] Y. J. Kim and J. S. Bolton, “Effects of rotation on the dynamics of a
circular cylindrical shell with applications to tire vibration, ” J. Sound
and vibration 275, 2003, pp. 605-621.
[13] Abaqus, Abaqus Documentation, Dassault Systmes, RI, USA, 2016.
422 ISNI:0000000091950263